* [patch 0/2] HFS: Convert bitmap_lock and extents_lock in mutexes
@ 2008-05-11 17:43 matthias
2008-05-11 17:43 ` [patch 1/2] HFS: Convert bitmap_lock in a mutex matthias
` (2 more replies)
0 siblings, 3 replies; 6+ messages in thread
From: matthias @ 2008-05-11 17:43 UTC (permalink / raw)
To: zippel; +Cc: linux-kernel, akpm
Apple Macintosh file system: The semaphores bitmap_lock and extents_lock are used as mutexes. Convert them to the mutex API
--
Matthias Kaehlcke
Embedded Linux Engineer
Barcelona
We build too many walls and not enough bridges
(Isaac Newton)
.''`.
using free software / Debian GNU/Linux | http://debian.org : :' :
`. `'`
gpg --keyserver pgp.mit.edu --recv-keys 47D8E5D4 `-
^ permalink raw reply [flat|nested] 6+ messages in thread* [patch 1/2] HFS: Convert bitmap_lock in a mutex 2008-05-11 17:43 [patch 0/2] HFS: Convert bitmap_lock and extents_lock in mutexes matthias @ 2008-05-11 17:43 ` matthias 2008-05-11 18:15 ` Matthias Kaehlcke 2008-05-11 17:43 ` [patch 2/2] HFS: Convert extents_lock " matthias 2008-05-11 18:06 ` [patch 0/2] HFS: Convert bitmap_lock and extents_lock in mutexes Matthias Kaehlcke 2 siblings, 1 reply; 6+ messages in thread From: matthias @ 2008-05-11 17:43 UTC (permalink / raw) To: zippel; +Cc: linux-kernel, akpm, Matthias Kaehlcke [-- Attachment #1: fs-hfs-bitmap_lock-to-mutex.diff --] [-- Type: text/plain, Size: 2568 bytes --] Apple Macintosh file system: The semaphore bitmap_lock is used as a mutex. Convert it to the mutex API Signed-off-by: Matthias Kaehlcke <matthias@kaehlcke.net> Index: hfs/bitmap.c =================================================================== --- hfs.orig/bitmap.c 2008-05-11 19:10:19.000000000 +0200 +++ hfs/bitmap.c 2008-05-11 19:11:17.000000000 +0200 @@ -145,7 +145,7 @@ if (!*num_bits) return 0; - down(&HFS_SB(sb)->bitmap_lock); + mutex_lock(&HFS_SB(sb)->bitmap_lock); bitmap = HFS_SB(sb)->bitmap; pos = hfs_find_set_zero_bits(bitmap, HFS_SB(sb)->fs_ablocks, goal, num_bits); @@ -162,7 +162,7 @@ HFS_SB(sb)->free_ablocks -= *num_bits; hfs_bitmap_dirty(sb); out: - up(&HFS_SB(sb)->bitmap_lock); + mutex_unlock(&HFS_SB(sb)->bitmap_lock); return pos; } @@ -205,7 +205,7 @@ if ((start + count) > HFS_SB(sb)->fs_ablocks) return -2; - down(&HFS_SB(sb)->bitmap_lock); + mutex_lock(&HFS_SB(sb)->bitmap_lock); /* bitmap is always on a 32-bit boundary */ curr = HFS_SB(sb)->bitmap + (start / 32); len = count; @@ -236,7 +236,7 @@ } out: HFS_SB(sb)->free_ablocks += len; - up(&HFS_SB(sb)->bitmap_lock); + mutex_unlock(&HFS_SB(sb)->bitmap_lock); hfs_bitmap_dirty(sb); return 0; Index: hfs/hfs_fs.h =================================================================== --- hfs.orig/hfs_fs.h 2008-05-11 19:10:40.000000000 +0200 +++ hfs/hfs_fs.h 2008-05-11 19:11:58.000000000 +0200 @@ -11,6 +11,7 @@ #include <linux/slab.h> #include <linux/types.h> +#include <linux/mutex.h> #include <linux/buffer_head.h> #include <linux/fs.h> @@ -139,7 +140,7 @@ struct nls_table *nls_io, *nls_disk; - struct semaphore bitmap_lock; + struct mutex bitmap_lock; unsigned long flags; Index: hfs/super.c =================================================================== --- hfs.orig/super.c 2008-05-11 19:10:34.000000000 +0200 +++ hfs/super.c 2008-05-11 19:12:13.000000000 +0200 @@ -372,7 +372,7 @@ sb->s_op = &hfs_super_operations; sb->s_flags |= MS_NODIRATIME; - init_MUTEX(&sbi->bitmap_lock); + mutex_init(&sbi->bitmap_lock); res = hfs_mdb_get(sb); if (res) { -- Matthias Kaehlcke Embedded Linux Engineer Barcelona We build too many walls and not enough bridges (Isaac Newton) .''`. using free software / Debian GNU/Linux | http://debian.org : :' : `. `'` gpg --keyserver pgp.mit.edu --recv-keys 47D8E5D4 `- ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [patch 1/2] HFS: Convert bitmap_lock in a mutex 2008-05-11 17:43 ` [patch 1/2] HFS: Convert bitmap_lock in a mutex matthias @ 2008-05-11 18:15 ` Matthias Kaehlcke 0 siblings, 0 replies; 6+ messages in thread From: Matthias Kaehlcke @ 2008-05-11 18:15 UTC (permalink / raw) To: zippel; +Cc: linux-kernel, akpm Apple Macintosh file system: The semaphore bitmap_lock is used as a mutex. Convert it to the mutex API Signed-off-by: Matthias Kaehlcke <matthias@kaehlcke.net> -- Index: linux-2.6/fs/hfs/bitmap.c =================================================================== --- linux-2.6.orig/fs/hfs/bitmap.c 2008-05-11 20:10:11.000000000 +0200 +++ linux-2.6/fs/hfs/bitmap.c 2008-05-11 20:10:18.000000000 +0200 @@ -145,7 +145,7 @@ if (!*num_bits) return 0; - down(&HFS_SB(sb)->bitmap_lock); + mutex_lock(&HFS_SB(sb)->bitmap_lock); bitmap = HFS_SB(sb)->bitmap; pos = hfs_find_set_zero_bits(bitmap, HFS_SB(sb)->fs_ablocks, goal, num_bits); @@ -162,7 +162,7 @@ HFS_SB(sb)->free_ablocks -= *num_bits; hfs_bitmap_dirty(sb); out: - up(&HFS_SB(sb)->bitmap_lock); + mutex_unlock(&HFS_SB(sb)->bitmap_lock); return pos; } @@ -205,7 +205,7 @@ if ((start + count) > HFS_SB(sb)->fs_ablocks) return -2; - down(&HFS_SB(sb)->bitmap_lock); + mutex_lock(&HFS_SB(sb)->bitmap_lock); /* bitmap is always on a 32-bit boundary */ curr = HFS_SB(sb)->bitmap + (start / 32); len = count; @@ -236,7 +236,7 @@ } out: HFS_SB(sb)->free_ablocks += len; - up(&HFS_SB(sb)->bitmap_lock); + mutex_unlock(&HFS_SB(sb)->bitmap_lock); hfs_bitmap_dirty(sb); return 0; Index: linux-2.6/fs/hfs/hfs_fs.h =================================================================== --- linux-2.6.orig/fs/hfs/hfs_fs.h 2008-05-11 20:10:11.000000000 +0200 +++ linux-2.6/fs/hfs/hfs_fs.h 2008-05-11 20:14:08.000000000 +0200 @@ -11,6 +11,7 @@ #include <linux/slab.h> #include <linux/types.h> +#include <linux/mutex.h> #include <linux/buffer_head.h> #include <linux/fs.h> @@ -139,7 +140,7 @@ struct nls_table *nls_io, *nls_disk; - struct semaphore bitmap_lock; + struct mutex bitmap_lock; unsigned long flags; Index: linux-2.6/fs/hfs/super.c =================================================================== --- linux-2.6.orig/fs/hfs/super.c 2008-05-11 20:10:11.000000000 +0200 +++ linux-2.6/fs/hfs/super.c 2008-05-11 20:10:18.000000000 +0200 @@ -372,7 +372,7 @@ sb->s_op = &hfs_super_operations; sb->s_flags |= MS_NODIRATIME; - init_MUTEX(&sbi->bitmap_lock); + mutex_init(&sbi->bitmap_lock); res = hfs_mdb_get(sb); if (res) { -- Matthias Kaehlcke Embedded Linux Engineer Barcelona La posibilidad de realizar un suenyo es lo que hace que la vida sea interesante .''`. using free software / Debian GNU/Linux | http://debian.org : :' : `. `'` gpg --keyserver pgp.mit.edu --recv-keys 47D8E5D4 `- ^ permalink raw reply [flat|nested] 6+ messages in thread
* [patch 2/2] HFS: Convert extents_lock in a mutex 2008-05-11 17:43 [patch 0/2] HFS: Convert bitmap_lock and extents_lock in mutexes matthias 2008-05-11 17:43 ` [patch 1/2] HFS: Convert bitmap_lock in a mutex matthias @ 2008-05-11 17:43 ` matthias 2008-05-11 18:16 ` Matthias Kaehlcke 2008-05-11 18:06 ` [patch 0/2] HFS: Convert bitmap_lock and extents_lock in mutexes Matthias Kaehlcke 2 siblings, 1 reply; 6+ messages in thread From: matthias @ 2008-05-11 17:43 UTC (permalink / raw) To: zippel; +Cc: linux-kernel, akpm, Matthias Kaehlcke [-- Attachment #1: fs-hfs-extents_lock-to-mutex.diff --] [-- Type: text/plain, Size: 3922 bytes --] Apple Macintosh file system: The semaphore extens_lock is used as a mutex. Convert it to the mutex API Signed-off-by: Matthias Kaehlcke <matthias@kaehlcke.net> Index: hfs/btree.c =================================================================== --- hfs.orig/btree.c 2008-05-11 19:28:16.000000000 +0200 +++ hfs/btree.c 2008-05-11 19:28:46.000000000 +0200 @@ -40,7 +40,7 @@ { struct hfs_mdb *mdb = HFS_SB(sb)->mdb; HFS_I(tree->inode)->flags = 0; - init_MUTEX(&HFS_I(tree->inode)->extents_lock); + mutex_init(&HFS_I(tree->inode)->extents_lock); switch (id) { case HFS_EXT_CNID: hfs_inode_read_fork(tree->inode, mdb->drXTExtRec, mdb->drXTFlSize, Index: hfs/extent.c =================================================================== --- hfs.orig/extent.c 2008-05-11 19:28:16.000000000 +0200 +++ hfs/extent.c 2008-05-11 19:29:43.000000000 +0200 @@ -343,16 +343,16 @@ goto done; } - down(&HFS_I(inode)->extents_lock); + mutex_lock(&HFS_I(inode)->extents_lock); res = hfs_ext_read_extent(inode, ablock); if (!res) dblock = hfs_ext_find_block(HFS_I(inode)->cached_extents, ablock - HFS_I(inode)->cached_start); else { - up(&HFS_I(inode)->extents_lock); + mutex_unlock(&HFS_I(inode)->extents_lock); return -EIO; } - up(&HFS_I(inode)->extents_lock); + mutex_unlock(&HFS_I(inode)->extents_lock); done: map_bh(bh_result, sb, HFS_SB(sb)->fs_start + @@ -375,7 +375,7 @@ u32 start, len, goal; int res; - down(&HFS_I(inode)->extents_lock); + mutex_lock(&HFS_I(inode)->extents_lock); if (HFS_I(inode)->alloc_blocks == HFS_I(inode)->first_blocks) goal = hfs_ext_lastblock(HFS_I(inode)->first_extents); else { @@ -425,7 +425,7 @@ goto insert_extent; } out: - up(&HFS_I(inode)->extents_lock); + mutex_unlock(&HFS_I(inode)->extents_lock); if (!res) { HFS_I(inode)->alloc_blocks += len; mark_inode_dirty(inode); @@ -487,7 +487,7 @@ if (blk_cnt == alloc_cnt) goto out; - down(&HFS_I(inode)->extents_lock); + mutex_lock(&HFS_I(inode)->extents_lock); hfs_find_init(HFS_SB(sb)->ext_tree, &fd); while (1) { if (alloc_cnt == HFS_I(inode)->first_blocks) { @@ -514,7 +514,7 @@ hfs_brec_remove(&fd); } hfs_find_exit(&fd); - up(&HFS_I(inode)->extents_lock); + mutex_unlock(&HFS_I(inode)->extents_lock); HFS_I(inode)->alloc_blocks = blk_cnt; out: Index: hfs/hfs_fs.h =================================================================== --- hfs.orig/hfs_fs.h 2008-05-11 19:28:17.000000000 +0200 +++ hfs/hfs_fs.h 2008-05-11 19:30:11.000000000 +0200 @@ -54,7 +54,7 @@ struct list_head open_dir_list; struct inode *rsrc_inode; - struct semaphore extents_lock; + struct mutex extents_lock; u16 alloc_blocks, clump_blocks; sector_t fs_blocks; Index: hfs/inode.c =================================================================== --- hfs.orig/inode.c 2008-05-11 19:28:17.000000000 +0200 +++ hfs/inode.c 2008-05-11 19:30:44.000000000 +0200 @@ -150,7 +150,7 @@ if (!inode) return NULL; - init_MUTEX(&HFS_I(inode)->extents_lock); + mutex_init(&HFS_I(inode)->extents_lock); INIT_LIST_HEAD(&HFS_I(inode)->open_dir_list); hfs_cat_build_key(sb, (btree_key *)&HFS_I(inode)->cat_key, dir->i_ino, name); inode->i_ino = HFS_SB(sb)->next_id++; @@ -281,7 +281,7 @@ HFS_I(inode)->flags = 0; HFS_I(inode)->rsrc_inode = NULL; - init_MUTEX(&HFS_I(inode)->extents_lock); + mutex_init(&HFS_I(inode)->extents_lock); INIT_LIST_HEAD(&HFS_I(inode)->open_dir_list); /* Initialize the inode */ -- Matthias Kaehlcke Embedded Linux Engineer Barcelona We build too many walls and not enough bridges (Isaac Newton) .''`. using free software / Debian GNU/Linux | http://debian.org : :' : `. `'` gpg --keyserver pgp.mit.edu --recv-keys 47D8E5D4 `- ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [patch 2/2] HFS: Convert extents_lock in a mutex 2008-05-11 17:43 ` [patch 2/2] HFS: Convert extents_lock " matthias @ 2008-05-11 18:16 ` Matthias Kaehlcke 0 siblings, 0 replies; 6+ messages in thread From: Matthias Kaehlcke @ 2008-05-11 18:16 UTC (permalink / raw) To: zippel; +Cc: linux-kernel, akpm Apple Macintosh file system: The semaphore extens_lock is used as a mutex. Convert it to the mutex API Signed-off-by: Matthias Kaehlcke <matthias@kaehlcke.net> -- Index: linux-2.6/fs/hfs/btree.c =================================================================== --- linux-2.6.orig/fs/hfs/btree.c 2008-05-11 20:11:54.000000000 +0200 +++ linux-2.6/fs/hfs/btree.c 2008-05-11 20:11:59.000000000 +0200 @@ -40,7 +40,7 @@ { struct hfs_mdb *mdb = HFS_SB(sb)->mdb; HFS_I(tree->inode)->flags = 0; - init_MUTEX(&HFS_I(tree->inode)->extents_lock); + mutex_init(&HFS_I(tree->inode)->extents_lock); switch (id) { case HFS_EXT_CNID: hfs_inode_read_fork(tree->inode, mdb->drXTExtRec, mdb->drXTFlSize, Index: linux-2.6/fs/hfs/extent.c =================================================================== --- linux-2.6.orig/fs/hfs/extent.c 2008-05-11 20:11:54.000000000 +0200 +++ linux-2.6/fs/hfs/extent.c 2008-05-11 20:11:59.000000000 +0200 @@ -343,16 +343,16 @@ goto done; } - down(&HFS_I(inode)->extents_lock); + mutex_lock(&HFS_I(inode)->extents_lock); res = hfs_ext_read_extent(inode, ablock); if (!res) dblock = hfs_ext_find_block(HFS_I(inode)->cached_extents, ablock - HFS_I(inode)->cached_start); else { - up(&HFS_I(inode)->extents_lock); + mutex_unlock(&HFS_I(inode)->extents_lock); return -EIO; } - up(&HFS_I(inode)->extents_lock); + mutex_unlock(&HFS_I(inode)->extents_lock); done: map_bh(bh_result, sb, HFS_SB(sb)->fs_start + @@ -375,7 +375,7 @@ u32 start, len, goal; int res; - down(&HFS_I(inode)->extents_lock); + mutex_lock(&HFS_I(inode)->extents_lock); if (HFS_I(inode)->alloc_blocks == HFS_I(inode)->first_blocks) goal = hfs_ext_lastblock(HFS_I(inode)->first_extents); else { @@ -425,7 +425,7 @@ goto insert_extent; } out: - up(&HFS_I(inode)->extents_lock); + mutex_unlock(&HFS_I(inode)->extents_lock); if (!res) { HFS_I(inode)->alloc_blocks += len; mark_inode_dirty(inode); @@ -487,7 +487,7 @@ if (blk_cnt == alloc_cnt) goto out; - down(&HFS_I(inode)->extents_lock); + mutex_lock(&HFS_I(inode)->extents_lock); hfs_find_init(HFS_SB(sb)->ext_tree, &fd); while (1) { if (alloc_cnt == HFS_I(inode)->first_blocks) { @@ -514,7 +514,7 @@ hfs_brec_remove(&fd); } hfs_find_exit(&fd); - up(&HFS_I(inode)->extents_lock); + mutex_unlock(&HFS_I(inode)->extents_lock); HFS_I(inode)->alloc_blocks = blk_cnt; out: Index: linux-2.6/fs/hfs/hfs_fs.h =================================================================== --- linux-2.6.orig/fs/hfs/hfs_fs.h 2008-05-11 20:11:54.000000000 +0200 +++ linux-2.6/fs/hfs/hfs_fs.h 2008-05-11 20:11:59.000000000 +0200 @@ -54,7 +54,7 @@ struct list_head open_dir_list; struct inode *rsrc_inode; - struct semaphore extents_lock; + struct mutex extents_lock; u16 alloc_blocks, clump_blocks; sector_t fs_blocks; Index: linux-2.6/fs/hfs/inode.c =================================================================== --- linux-2.6.orig/fs/hfs/inode.c 2008-05-11 20:11:54.000000000 +0200 +++ linux-2.6/fs/hfs/inode.c 2008-05-11 20:11:59.000000000 +0200 @@ -150,7 +150,7 @@ if (!inode) return NULL; - init_MUTEX(&HFS_I(inode)->extents_lock); + mutex_init(&HFS_I(inode)->extents_lock); INIT_LIST_HEAD(&HFS_I(inode)->open_dir_list); hfs_cat_build_key(sb, (btree_key *)&HFS_I(inode)->cat_key, dir->i_ino, name); inode->i_ino = HFS_SB(sb)->next_id++; @@ -281,7 +281,7 @@ HFS_I(inode)->flags = 0; HFS_I(inode)->rsrc_inode = NULL; - init_MUTEX(&HFS_I(inode)->extents_lock); + mutex_init(&HFS_I(inode)->extents_lock); INIT_LIST_HEAD(&HFS_I(inode)->open_dir_list); /* Initialize the inode */ -- Matthias Kaehlcke Embedded Linux Engineer Barcelona Don't walk behind me, I may not lead Don't walk in front of me, I may not follow Just walk beside me and be my friend (Albert Camus) .''`. using free software / Debian GNU/Linux | http://debian.org : :' : `. `'` gpg --keyserver pgp.mit.edu --recv-keys 47D8E5D4 `- ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [patch 0/2] HFS: Convert bitmap_lock and extents_lock in mutexes 2008-05-11 17:43 [patch 0/2] HFS: Convert bitmap_lock and extents_lock in mutexes matthias 2008-05-11 17:43 ` [patch 1/2] HFS: Convert bitmap_lock in a mutex matthias 2008-05-11 17:43 ` [patch 2/2] HFS: Convert extents_lock " matthias @ 2008-05-11 18:06 ` Matthias Kaehlcke 2 siblings, 0 replies; 6+ messages in thread From: Matthias Kaehlcke @ 2008-05-11 18:06 UTC (permalink / raw) To: zippel; +Cc: linux-kernel, akpm El Sun, May 11, 2008 at 07:43:31PM +0200 matthias@kaehlcke.net ha dit: > Apple Macintosh file system: The semaphores bitmap_lock and extents_lock are used as mutexes. Convert them to the mutex API sorry, i just realized that the patches use relative paths. i'm going to submit a fix -- Matthias Kaehlcke Embedded Linux Engineer Barcelona Usually when people are sad, they don't do anything. They just cry over their condition. But when they get angry, they bring about a change (Malcolm X) .''`. using free software / Debian GNU/Linux | http://debian.org : :' : `. `'` gpg --keyserver pgp.mit.edu --recv-keys 47D8E5D4 `- ^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2008-05-11 18:16 UTC | newest] Thread overview: 6+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2008-05-11 17:43 [patch 0/2] HFS: Convert bitmap_lock and extents_lock in mutexes matthias 2008-05-11 17:43 ` [patch 1/2] HFS: Convert bitmap_lock in a mutex matthias 2008-05-11 18:15 ` Matthias Kaehlcke 2008-05-11 17:43 ` [patch 2/2] HFS: Convert extents_lock " matthias 2008-05-11 18:16 ` Matthias Kaehlcke 2008-05-11 18:06 ` [patch 0/2] HFS: Convert bitmap_lock and extents_lock in mutexes Matthias Kaehlcke
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox